CN101902487B - Queue scheduling method and device based on linked list - Google Patents
Queue scheduling method and device based on linked list Download PDFInfo
- Publication number
- CN101902487B CN101902487B CN2009100856317A CN200910085631A CN101902487B CN 101902487 B CN101902487 B CN 101902487B CN 2009100856317 A CN2009100856317 A CN 2009100856317A CN 200910085631 A CN200910085631 A CN 200910085631A CN 101902487 B CN101902487 B CN 101902487B
- Authority
- CN
- China
- Prior art keywords
- queue
- list
- formation
- priority
- queuing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
Abstract
The invention discloses a queue scheduling method based on a linked list, which comprises the following steps: setting the number of addresses in a queue linked list to be not less than the number of queues, and dividing the queue linked list into different queue sub-linked lists according to the priorities of the queues, wherein the number of the addresses in the queue sub-linked lists is not less than the number of all queues under the priorities corresponding to the queue sub-linked lists; setting a queue linked list identifier for each queue to identify whether each queue is queued in the queue linked list or not; before a queue meeting the queue conditions is added into the queue linked list, judging whether the queue is queued in the queue linked list or not according to the queue linked list identifier of the queue; if the queue is queued, doing not carry out the adding treatment; and if the queue is not queued, adding the queue into the tail of the queue sub-linked list corresponding to the queue priority, and changing the queue linked list identifier of the queue into an identification for showing that the queue is queued in the queue linked list. The invention also discloses a queue scheduling device based on the linked list. The invention ensures the fairness when the queues with the same priority are scheduled.
Description
Technical field
The present invention relates to the queue scheduling technology, relate in particular to a kind of array dispatching method and device that mixes based on the multiple dispatching technique of chained list.
Background technology
Queue scheduling, i.e. formation trigger to check (such as receiving grant triggers etc.) by certain, the arbitration formation go out group condition, according to certain rule-based scheduling formation orderly go out team.Dispatching algorithm commonly used mainly contains poll (RR, Round Robin) dispatching algorithm and strict priority (SP, Strict Priority) dispatching algorithm.
The realization principle of polling dispatching algorithm be in a certain order one by one, the cyclic polling formation goes out team.Go out team in formation of each clock period scheduling, then go out team at the next queue scheduling of next clock period poll.And the strict priority scheduling algorithm is to set different priority to formation, and the formation of at every turn dispatching limit priority goes out team, and the formation of low priority is not only satisfied at high-priority queue just can obtain out team's scheduling after group condition.
At present, during for multipriority, many formations output scheduling, common disposal route is with a first-in first-out (FIFO to the formation of equal priority, First Input First Output) the buffer memory queuing is carried out in formation, then adopts the output of SP strict priority scheduling between different priorities queuing FIFO.
Because being the advanced person, the principle of FIFO obtains first service, same queue can enter the FIFO queuing repeatedly, when a certain formation flow is larger, will repeatedly appear among the FIFO and line up, this will so that in the same priority formation a very long time of low discharge all can not get service, only have the by the time formation of high flow capacity repeatedly to go out team and just can take turns to many times.This just can not embody for the same priority formation and should go out the thought of team by fair poll, has caused team's injustice.Another problem is when formation is more, and the risk that FIFO can occur overflowing for avoiding occurring this phenomenon, needs the larger fifo queue of configuration, and this will expend quite huge RAM resource.Going out the more situation of group number of queues and queue priority such as number of queues is 32K, queue priority is 6, for avoiding the appearance of spillover, going out team's queuing is the FIFO of 32K with regard to needing 6 degree of depth at least, this has expended a large amount of RAM resources undoubtedly, but a big chunk in the above-mentioned RAM resource that disposes can not be used, and has also caused the serious wasting of resources.
Summary of the invention
In view of this, fundamental purpose of the present invention is to provide a kind of array dispatching method based on chained list and device, when guaranteeing the strict priority scheduling algorithm, can guarantee that the formation of same priority is dispatched liberally.
For achieving the above object, technical scheme of the present invention is achieved in that
A kind of array dispatching method based on chained list comprises:
The number of addresses that arranges in the queue list is no less than number of queues, and by priority under the formation described queue list is divided into different queuing child lists, wherein, the number of addresses in the described queuing child list be no less than described queuing child list all number of queues under the corresponding priority; For individual queue arranges its queue list identifier of whether lining up of sign in described queue list; And
Before the formation that will meet queue conditions inserts described queue list, judge according to the queue list identifier of described formation whether described formation lines up in described queue list, then do not carry otu insertion process if lined up, if queuing does not then insert described formation the chained list afterbody of queuing child list corresponding to described queue priority, and the queue list identifier of described formation is revised as the sign of having lined up in described queue list.
Preferably, described method also comprises:
Select the highest nonempty queue child list of priority to carry out queue scheduling in described queue list, the nonempty queue child list that the nonempty queue child list that priority is the highest is dispatched complete rear selection time priority carries out queue scheduling.
Preferably, the queuing child list is when carrying out queue scheduling, begun by the head of chained list, carries out queue scheduling by the indication of chain list index.
Preferably, described method also comprises: when the formation that goes out team that priority is identical and formation to be joined the team all met queue conditions, the formation that preferentially will go out team inserted in the described queue list.
Preferably, described queue list identifier is 1 bit, and 0 sign is lined up in described queue list, and 1 sign is not lined up in described queue list; Perhaps, 0 sign is not lined up in described queue list, and 1 sign is lined up in described queue list.
Preferably, each queue number is corresponding to an address in the queue list.
A kind of queue scheduling device based on chained list comprises:
Setting unit, the address that is used for arranging queue list is no less than the number of total formation, and, for individual queue arranges its queue list identifier of whether lining up of sign in described queue list;
Division unit is used for by priority under the formation described queue list being divided into different queuing child lists, wherein, the address in the described queuing child list be no less than described queuing child list all number of queues under the corresponding priority;
Judging unit, be used for before the formation that will meet queue conditions inserts described queue list, judge according to the queue list identifier of described formation whether described formation lines up, does not then carry otu insertion process if lined up in described queue list, do not insert the unit if line up then to trigger;
Insert the unit, be used for described formation is inserted the chained list afterbody of queuing child list corresponding to described queue priority, and trigger sign and revise the unit; And
Sign is revised the unit, is used for the queue list identifier of described formation is revised as the sign of having lined up at described queue list.
Preferably, described device also comprises:
Scheduling unit is used for selecting the highest queuing child list of priority to carry out queue scheduling at described queue list, and the queuing child list that priority is the highest is dispatched the complete rear selection time the highest queuing child list of priority and carried out queue scheduling.
Preferably, when described scheduling unit queuing child list carries out queue scheduling, begun by the head of chained list, carry out queue scheduling by the indication of chain list index.
Preferably, the described unit that inserts preferentially will go out right formation and insert in the described queue list when identical the going out right formation and enter right formation of priority all meets queue conditions.
Preferably, each queue number is corresponding to an address in the queue list.
Among the present invention, the number of addresses of queue list is set by number of queues, make the number of addresses of queue list be no less than number of queues, priority by formation arranges different queuing child lists in queue list again, formation enter to the time, press queue priority and select corresponding queuing child list, before inserting the queuing child list, judge that whether this formation lines up, if queuing then no longer inserts in the queuing child list, otherwise inserts in the queuing child list.During scheduling, in queue list, choose the highest nonempty queue child list of priority, from the head of chained list, carry out queue scheduling by the indication of chain list index, after queue scheduling in the selected nonempty queue child list was complete, the nonempty queue child list of choosing secondary priority carried out queue scheduling.The present invention dispatches by the mode of first-in first-out when processing the same priority formation, but for for the formation of lining up in the queue list, can not be inserted in the queue list fairness when having guaranteed like this same priority queue scheduling again.
Description of drawings
Fig. 1 is the process flow diagram that the present invention is based on the array dispatching method of chained list;
Fig. 2 formation of the present invention inserts the structural representation of queuing child list;
Fig. 3 is the composition structural representation that the present invention is based on the queue scheduling device of chained list.
Embodiment
Basic thought of the present invention is: the number of addresses that queue list is set by number of queues, make the number of addresses of queue list be no less than number of queues, priority by formation arranges different queuing child lists in queue list again, formation enter to the time, press queue priority and select corresponding queuing child list, before inserting the queuing child list, judge whether this formation lines up in the queuing child list, if queuing then no longer inserts in the queuing child list, otherwise inserts.During scheduling, in queue list, choose the highest nonempty queue child list of priority, from the head of chained list, carry out queue scheduling by the indication of chain list index, after queue scheduling in the selected nonempty queue child list was complete, the nonempty queue child list of choosing secondary priority carried out queue scheduling.The present invention dispatches by the mode of first-in first-out when processing the same priority formation, but for for the formation of lining up in the queue list, can not be inserted in the queue list fairness when having guaranteed like this same priority queue scheduling again.
For making the purpose, technical solutions and advantages of the present invention clearer, by the following examples and with reference to accompanying drawing, the present invention is described in more detail.
Fig. 1 is the process flow diagram that the present invention is based on the array dispatching method of chained list, and as shown in Figure 1, the array dispatching method that the present invention is based on chained list may further comprise the steps:
Step 101: the number of addresses that arranges in the queue list is no less than number of queues, and by priority under the formation described queue list is divided into different queuing child lists, wherein, the number of addresses in the described queuing child list be no less than described queuing child list all number of queues under the corresponding priority; For individual queue arranges its queue list identifier of whether lining up of sign in described queue list.
The number of addresses of queue list is set according to the number of all formations in the system, 32K formation is as example in the aforementioned background art, the number of addresses that queue list then is set is at least 32K, the address of queue list needs at least, and the character of 15bit carries, be the address that each formation need have the queue list of a correspondence, so that the address of depositing this formation to be arranged when formation is carried otu insertion process.For queue list, difference by queue priority is set to different queuing child lists again, so that the formation of respective priority is ranked, the formation of respective priority is inserted in the queuing child list of corresponding priority, number of addresses in the queuing child list can not be less than all number of queues under this priority, so that all formations under this priority all can be ranked in corresponding queuing child list.Among the present invention, formation mainly refers to the affiliated different data queue of user.Best mode is that the number of addresses in the queuing child list is corresponding one by one with all formations.
Simultaneously, for individual queue arranges its queue list identifier of whether lining up of sign in described queue list, this queue list identifier is the character of 1 bit, and wherein, 0 sign is lined up in described queue list, and 1 sign is not lined up in described queue list; Perhaps, 0 sign is not lined up in described queue list, and 1 sign is lined up in described queue list.Among the present invention, the queue list identifier address of formation is corresponding with the sequence number of formation, with queue list identifier corresponding to easy-to-look-up formation.
Step 102: before the formation that will meet queue conditions inserts described queue list, judge according to the queue list identifier of described formation whether described formation lines up in described queue list, then do not carry otu insertion process if lined up, if queuing does not then insert described formation the chained list afterbody of queuing child list corresponding to described queue priority, and the queue list identifier of described formation is revised as the sign of having lined up in described queue list.
After setting up queue list, judge that namely which formation meets queue conditions, for the formation that meets queue conditions, it is inserted in the queue list, concrete, before inserting queue list, need to determine the priority of formation, then, check again the queue list identifier of this formation, determine whether this formation lines up in queue list, if line up, then no longer process this formation, namely no longer this formation is inserted in the queue list, if and in queue list, do not line up, then this formation is inserted in the queuing child list corresponding with this queue priority.Below describe in detail by concrete example.
Suppose that the queue number of joining the team successively is: 35,13,0,8,0,6,1,64,0,3,55,30,20,80,9,101 ... etc.Wherein, 0,1, No. 3 formations belong to the first priority (limit priority), and 6,8,9 belong to the second priority, and 13,20,30,35 belong to the 3rd priority, and 80,64,55,101 belong to the 4th priority.Suppose that each formation all satisfies queue conditions.No. 35 formations are joined the team at first, (the queue list identifier address is fully corresponding with the formation sequence for inquiry queue list identifier address 35, this is with very convenient searching formation queue list identifier) content, discovery is 0 when (suppose 0 sign formation do not line up) in queue list, No. 35 formations are inserted in the queue list, simultaneously, the queue list identifier of this address is revised as 1 (the sign formation is lined up) in queue list.Inquiry queue list identifier address 35 is 1 o'clock, just shows that this formation lines up in chained list, no longer processes joining the team of this formation.According to the order of joining the team, No. 13, the formation of back, No. 0, No. 8, No. 0 grade is inquired about the queue list identifier successively.When the queue list identifier is inquired about in No. 0 formation for the second time, found in the chained list the inside, will no longer it have been inserted in the queue list.
The number of addresses of queue list is corresponding with queue number (number of queues), and each formation all has the address of queuing.The present invention is that the priority by formation inserts this formation in the queuing child list of corresponding priority when inserting formation in the queue list.Still take the formation of aforementioned numbering as example, will insert successively queue number in the queuing child list of the first priority is No. 0, No. 1, and No. 3 formation.Fig. 2 formation of the present invention inserts the structural representation of queuing child list, as shown in Figure 2, queuing child list address 0 is exactly the chained list first address, what deposit is No. 0 queue number, what deposit in No. 1 address is No. 1 queue number, and what deposit in No. 3 addresses is No. 3 formations, owing to only inserted 3 formations in the queuing child list, the next hop address of address 3 pointers is exactly chained list tail address, and content is empty.In the time of in formation being inserted the queuing child list, begin to deposit formation from the tail address of queuing child list, aforesaid row number are No. 0, No. 1, when No. 3 formation inserts the queuing child list, when inserting for No. 0, because the queuing child list of the first priority is empty at this moment, address 0 is exactly chained list tail address, with No. 0 queue stores in address 0, at this moment, address 1 is exactly chained list tail address, and No. 1 formation is inserted in the address 1 again.As shown in Figure 2, formation such as aforementioned 8,6, No. 9 formations for the second priority, still insert in the queuing child list by preceding method, queuing child list for the second priority, first address can be address 8, and the down hop of address 8 is address 6, stores No. 6 formations, and the down hop of address 6 is address 9, is used for No. 9 formations of storage.The formation of storage gets final product in strict accordance with the address pointer in the queuing child list, the first formation that inserts of first address storage, and afterwards all formations are sequentially stored in the tail address of queuing child list.
The implementation of optimum of the present invention is that number of addresses is to equate with number of queues, and at this moment, number of queues and number of addresses are corresponding one by one.Among the present invention, as long as avoid formation corresponding to same address.
The present invention only inserts the queuing child list take the formation of first and second priority and is illustrated as example, and it is identical therewith that the formation of all the other priority inserts the mode of queuing child list of respective priority, repeats no more here.
The present invention also comprises: select the highest nonempty queue child list of priority to carry out queue scheduling in described queue list, the nonempty queue child list that the nonempty queue child list that priority is the highest is dispatched complete rear selection time priority carries out queue scheduling.When in the queuing child list, carrying out queue scheduling, at first from the head output of queuing child list, dispatch next formation by the indication of queuing child list head pointer again and go out team, until the queuing child list is empty.
Still take aforesaid formation as example, sequentially be followed successively by No. 0 No. 1 such as the output of the first priority query, No. 3 ..., after the queue scheduling in queuing child list corresponding to the first priority is complete, find out the highest nonempty queue child list of current priority, the second priority queueing child list in this way, its formation output is followed successively by No. 8, No. 6, No. 9 ..., then be the 3rd priority queueing child list, its formation output is followed successively by No. 13 No. 35, No. 30, No. 20, etc.
Fig. 3 is the composition structural representation that the present invention is based on the queue scheduling device of chained list, as shown in Figure 3, the queue scheduling device that the present invention is based on chained list comprises setting unit 30, division unit 31, judging unit 32, inserts unit 33 and sign modification unit 34, wherein, setting unit 30 is no less than the number of total formation for the address that queue list is set, and, for individual queue arranges its queue list identifier of whether lining up of sign in described queue list; Division unit 31 is used for by priority under the formation described queue list being divided into different queuing child lists, wherein, the address in the described queuing child list be no less than described queuing child list all number of queues under the corresponding priority; Judging unit 32 was used for before the formation that will meet queue conditions inserts described queue list, judge according to the queue list identifier of described formation whether described formation lines up in described queue list, then do not carry otu insertion process if lined up, do not insert unit 33 if line up then to trigger; Insert the chained list afterbody that unit 33 is used for described formation is inserted queuing child list corresponding to described queue priority, and trigger sign and revise unit 34; Sign is revised unit 34 and is used for the queue list identifier of described formation is revised as the sign of having lined up at described queue list.
As shown in Figure 3, the queue scheduling device that the present invention is based on chained list also comprises scheduling unit 35, be used for selecting the highest queuing child list of priority to carry out queue scheduling at described queue list, the queuing child list that priority is the highest is dispatched the complete rear selection time the highest queuing child list of priority and is carried out queue scheduling.。
It will be appreciated by those skilled in the art that scheduling unit 35 is to arrange for technical scheme that optimization the present invention is based on the queue scheduling device of chained list, is not the necessary parts of realizing technical solution of the present invention.
Those skilled in the art are to be understood that, the present invention's queue scheduling device based on chained list shown in Figure 3 is for realizing that the array dispatching method based on chained list shown in Figure 1 designs, each processing unit in the device shown in Figure 3 and the practical function of processing module can be understood with reference to the associated description in the method shown in Figure 1, the function of each unit and each module can realize by the program that runs on the processor, also can realize by corresponding logical circuit.
The above is preferred embodiment of the present invention only, is not for limiting protection scope of the present invention.
Claims (11)
1. the array dispatching method based on chained list is characterized in that, comprising:
The number of addresses that arranges in the queue list is no less than number of queues, and by priority under the formation described queue list is divided into different queuing child lists, wherein, the number of addresses in the described queuing child list be no less than described queuing child list all number of queues under the corresponding priority; For individual queue arranges its queue list identifier of whether lining up of sign in described queue list; And
Before the formation that will meet queue conditions inserts described queue list, judge according to the queue list identifier of described formation whether described formation lines up in described queue list, then do not carry otu insertion process if lined up, if queuing does not then insert described formation the chained list afterbody of queuing child list corresponding to described queue priority, and the queue list identifier of described formation is revised as the sign of having lined up in described queue list.
2. method according to claim 1 is characterized in that, described method also comprises:
Select the highest nonempty queue child list of priority to carry out queue scheduling in described queue list, the nonempty queue child list that the nonempty queue child list that priority is the highest is dispatched complete rear selection time priority carries out queue scheduling.
3. method according to claim 2 is characterized in that, the queuing child list is when carrying out queue scheduling, is begun by the head of chained list, carries out queue scheduling by the indication of chain list index.
4. method according to claim 1 is characterized in that, described method also comprises: when the formation that goes out team that priority is identical and formation to be joined the team all met queue conditions, the formation that preferentially will go out team inserted in the described queue list.
5. method according to claim 1 is characterized in that, described queue list identifier is 1 bit, and 0 sign is lined up in described queue list, and 1 sign is not lined up in described queue list; Perhaps, 0 sign is not lined up in described queue list, and 1 sign is lined up in described queue list.
6. method according to claim 1 is characterized in that, each queue number is corresponding to an address in the queue list.
7. the queue scheduling device based on chained list is characterized in that, comprising:
Setting unit, the address that is used for arranging queue list is no less than the number of total formation, and, for individual queue arranges its queue list identifier of whether lining up of sign in described queue list;
Division unit is used for by priority under the formation described queue list being divided into different queuing child lists, wherein, the address in the described queuing child list be no less than described queuing child list all number of queues under the corresponding priority;
Judging unit, be used for before the formation that will meet queue conditions inserts described queue list, judge according to the queue list identifier of described formation whether described formation lines up, does not then carry otu insertion process if lined up in described queue list, do not insert the unit if line up then to trigger;
Insert the unit, be used for described formation is inserted the chained list afterbody of queuing child list corresponding to described queue priority, and trigger sign and revise the unit; And
Sign is revised the unit, is used for the queue list identifier of described formation is revised as the sign of having lined up at described queue list.
8. device according to claim 7 is characterized in that, described device also comprises:
Scheduling unit is used for selecting the highest queuing child list of priority to carry out queue scheduling at described queue list, and the queuing child list that priority is the highest is dispatched the complete rear selection time the highest queuing child list of priority and carried out queue scheduling.
9. device according to claim 8 is characterized in that, when described scheduling unit queuing child list carries out queue scheduling, is begun by the head of chained list, carries out queue scheduling by the indication of chain list index.
10. device according to claim 7 is characterized in that, the described unit that inserts is when the identical formation that goes out team of priority all meets queue conditions with the formation of joining the team, and the formation that preferentially will go out team inserts in the described queue list.
11. device according to claim 7 is characterized in that, each queue number is corresponding to an address in the queue list.
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100856317A CN101902487B (en) | 2009-05-26 | 2009-05-26 | Queue scheduling method and device based on linked list |
| US13/258,936 US8463967B2 (en) | 2009-05-26 | 2010-03-24 | Method and device for scheduling queues based on chained list |
| EP10780008.8A EP2437166A4 (en) | 2009-05-26 | 2010-03-24 | Method and device for scheduling queues based on chained list |
| PCT/CN2010/071225 WO2010135926A1 (en) | 2009-05-26 | 2010-03-24 | Method and device for scheduling queues based on chained list |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100856317A CN101902487B (en) | 2009-05-26 | 2009-05-26 | Queue scheduling method and device based on linked list |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101902487A CN101902487A (en) | 2010-12-01 |
| CN101902487B true CN101902487B (en) | 2013-03-20 |
Family
ID=43222140
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100856317A Expired - Fee Related CN101902487B (en) | 2009-05-26 | 2009-05-26 | Queue scheduling method and device based on linked list |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US8463967B2 (en) |
| EP (1) | EP2437166A4 (en) |
| CN (1) | CN101902487B (en) |
| WO (1) | WO2010135926A1 (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2725750B1 (en) | 2011-08-04 | 2017-11-08 | Huawei Technologies Co., Ltd. | Queue scheduling method and device |
| CN103546392B (en) * | 2012-07-12 | 2017-10-10 | 中兴通讯股份有限公司 | Single queue cycle dispatching method and device |
| US10548150B2 (en) * | 2013-03-14 | 2020-01-28 | Federated Wireless, Inc. | Radio resource management |
| JP6241128B2 (en) * | 2013-08-14 | 2017-12-06 | 富士通株式会社 | Communication apparatus and packet scheduling method |
| US9178889B2 (en) * | 2013-09-27 | 2015-11-03 | Paypal, Inc. | Systems and methods for pairing a credential to a device identifier |
| CN104753818B (en) * | 2013-12-30 | 2019-04-30 | 深圳市中兴微电子技术有限公司 | A kind of array dispatching method and device |
| CN106325758B (en) * | 2015-06-17 | 2019-10-22 | 深圳市中兴微电子技术有限公司 | A kind of queue storage space management method and device |
| CN104965686B (en) * | 2015-07-20 | 2019-01-22 | 北京金山安全软件有限公司 | Priority queue design method and priority queue design device |
| CN107040478A (en) * | 2015-07-29 | 2017-08-11 | 深圳市中兴微电子技术有限公司 | Token application processing method and device |
| CN111277900B (en) * | 2018-12-05 | 2022-12-23 | 深圳市茁壮网络股份有限公司 | Starting method and device of set top box |
| CN111124355B (en) * | 2019-12-12 | 2023-04-07 | 东软集团股份有限公司 | Information processing method and device, readable storage medium and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1322091A (en) * | 2000-03-02 | 2001-11-14 | 阿尔卡塔尔互联网运行公司 | Qualifid priority queue dispatcher |
| US6661802B1 (en) * | 1998-10-27 | 2003-12-09 | Fujitsu Network Communications, Inc. | Congestion management |
| US20040213243A1 (en) * | 2001-01-16 | 2004-10-28 | Ying-Chien Lin | Transmission components for processing VLAN tag and priority packets supported by using single chip's buffer structure |
| CN1979424A (en) * | 2005-11-29 | 2007-06-13 | 腾讯科技（深圳）有限公司 | Method for managing priority queue |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5247677A (en) * | 1992-05-22 | 1993-09-21 | Apple Computer, Inc. | Stochastic priority-based task scheduler |
| US5765165A (en) * | 1996-02-29 | 1998-06-09 | Sun Microsystems, Inc. | Fast method of determining duplicates on a linked list |
| US7302686B2 (en) * | 2001-07-04 | 2007-11-27 | Sony Corporation | Task management system |
| US20050047425A1 (en) | 2003-09-03 | 2005-03-03 | Yonghe Liu | Hierarchical scheduling for communications systems |
| US7370326B2 (en) * | 2004-04-02 | 2008-05-06 | Emulex Design & Manufacturing Corporation | Prerequisite-based scheduler |
| US7555579B2 (en) * | 2004-05-21 | 2009-06-30 | Nortel Networks Limited | Implementing FIFOs in shared memory using linked lists and interleaved linked lists |
| US7474662B2 (en) * | 2005-04-29 | 2009-01-06 | International Business Machines Corporation | Systems and methods for rate-limited weighted best effort scheduling |
| US7822915B2 (en) * | 2007-06-30 | 2010-10-26 | Alcatel-Lucent Usa Inc. | Memory controller for packet applications |
-
2009
- 2009-05-26 CN CN2009100856317A patent/CN101902487B/en not_active Expired - Fee Related
-
2010
- 2010-03-24 WO PCT/CN2010/071225 patent/WO2010135926A1/en active Application Filing
- 2010-03-24 US US13/258,936 patent/US8463967B2/en active Active
- 2010-03-24 EP EP10780008.8A patent/EP2437166A4/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6661802B1 (en) * | 1998-10-27 | 2003-12-09 | Fujitsu Network Communications, Inc. | Congestion management |
| CN1322091A (en) * | 2000-03-02 | 2001-11-14 | 阿尔卡塔尔互联网运行公司 | Qualifid priority queue dispatcher |
| US20040213243A1 (en) * | 2001-01-16 | 2004-10-28 | Ying-Chien Lin | Transmission components for processing VLAN tag and priority packets supported by using single chip's buffer structure |
| CN1979424A (en) * | 2005-11-29 | 2007-06-13 | 腾讯科技（深圳）有限公司 | Method for managing priority queue |
Also Published As
| Publication number | Publication date |
|---|---|
| US8463967B2 (en) | 2013-06-11 |
| EP2437166A1 (en) | 2012-04-04 |
| US20120017055A1 (en) | 2012-01-19 |
| EP2437166A4 (en) | 2013-04-24 |
| CN101902487A (en) | 2010-12-01 |
| WO2010135926A1 (en) | 2010-12-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101902487B (en) | Queue scheduling method and device based on linked list | |
| CN101694631B (en) | Real time job dispatching patcher and method | |
| US7876763B2 (en) | Pipeline scheduler including a hierarchy of schedulers and multiple scheduling lanes | |
| CN104268018B (en) | Job scheduling method and job scheduler in a kind of Hadoop clusters | |
| CN112214319B (en) | Task scheduling method for sensing computing resources | |
| CN106293918A (en) | A kind of dispatch the method for process, system and computer | |
| CN102497310A (en) | Service flow shaping method and device for cascaded port of avionics full duplex switched Ethernet (AFDX) switch | |
| CN101072176A (en) | Report processing method and system | |
| CN109840145B (en) | Multistage scheduling method, device, network chip and storage medium | |
| US20060274779A1 (en) | Filling token buckets of schedule entries | |
| CN111274021B (en) | GPU cluster task scheduling and distributing method | |
| CN110515713A (en) | A kind of method for scheduling task, equipment and computer storage medium | |
| US7565496B2 (en) | Sharing memory among multiple information channels | |
| CN100456744C (en) | Data dispatching method and system | |
| CN108768865B (en) | Method for generating time-triggered service scheduling table supporting multicast service | |
| US20080080504A1 (en) | System and method for managing flow of a plurality of packets in a lossless communication network | |
| CN103546392B (en) | Single queue cycle dispatching method and device | |
| CN109800074A (en) | Task data concurrently executes method, apparatus and electronic equipment | |
| CN1618215A (en) | Method for real time network traffic admission and scheduling | |
| CN111124716A (en) | Multi-priority queue message transmission method in unmanned aerial vehicle | |
| CN109347764A (en) | Scheduling method, system and medium for realizing bandwidth matching | |
| CN116089033A (en) | Task scheduling method based on multistage heterogeneous dynamic queue | |
| CN102594670B (en) | Multiport multi-flow scheduling method, device and equipment | |
| CN113170500B (en) | Information transmission method and device | |
| CN101610552A (en) | The dispatching method of common source and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130320 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |